Text Miner Description

is a software tool that can extract hidden knowledge from legacy texts. The Text Miner is ideal to create terminology lists, taxonomies, glossaries, Knowledge Management and eLearning topics. These lists are valuable for Controlled English, Simplified Technical English, Plain English, Machine Translation, Help Desks, Call Center scripts, QA/QC procedures, Technical Training curriculums and eDiscovery Litigation Support.

How the Text Miner works

The user gathers files for a subject domain, for example, an aircraft hydraulic system. These files can be in Word, XML, HTML or .pdf format. The SMART Text Miner reads these files and creates a report that shows the terms in context with frequency and origin. The user reviews the reports with Subject Matter Experts (SMEs) to form two lists, valid terminology and synonyms. The approved terms are entered into the SMART Lexicon Manager for terminology management and used in the MAXit Checker.

Main Screen

This screen lets the user add Text Miner report titles.

Lexicon Manager - Screen No. 1
Contact Smart  Questions welcomed.

Related Topics



Text Miner Features

  • An easy-to-use point and click interface.
  • A knowledge of linguistics is not needed.
  • A STOP LIST removes common terms.
  • Ability to link to the Lexicon Manager.
  • Multiple files can be processed and combined into one report.
  • Multiple text formats can be processed.
  • A left and right context shows use.
  • Mining-at-a-Glance report mode.
  • Automatic functions to harvest terms.
  • User can set the number of occurrences.
  • The frequency counter shows the importance of each term in context.
  • All graphics are ignored automatically.
  • All format tags are skipped.
  • Software can be programmed to skip fields in input text, example, email headers.
  • Output reports in .pdf format are shared.
  • Ability to assign part of speech codes.
  • An interactive view mode avoids paper.
  • Results can standardize nomenclature.
  • Reports can discover trends through terms.
  • Reports can show terms required for regulatory agencies, example, warnings.
  • On-line help with linguistic examples.
  • Ideal to find valid abbreviations, labels, computer messages, non-English terms, trademarks, signal names, acronyms, marks, geographical and proper names.
  • Reports help eLearning tool developers.
  • Reports can show synonyms for SMART Controlled English dictionaries.
  • Custom reports with headers and footers.
  • Reports help translators build word lists.
  • Text mining takes days, not months.